Correspondence
Serum cancer antigen 15-3 concentrations in patients with sickle cell disease
Cancer antigen (CA) 15-3, which is associated with breast carcinoma, is an epithelial mucin and a product of the MUC1 gene (MUC1) (Hayes et al, 1986). Elevated levels of CA 15-3 also occur in patients with benign conditions such as inflammatory disease (Colomer et al, 1989; Valerio Marzano et al, 1998). Sickle cell disease (SCD) is an inflammatory disorder characterised by chronic haemolysis, endothelial activation and vaso-occlusion (Hebbel et al, 2004). Recent evidence showed that MUC1 was also expressed in the haematopoietic lineages (Rughetti et al, 2003), which prompted us to examine the possible association of CA 15-3 with SCD. Fifty-one patients with homozygous SCD, 15 patients with previously treated active metastatic breast carcinoma, and 20 healthy volunteers, sex and age-matched, were enrolled in the study. The SCD patients were divided into two subgroups: individuals with painful vaso-occlusive crises (n ¼ 17) and those with steady-state disease (n ¼ 34). Patients with active infection, cirrhosis, severe organ dysfunction, cancer, pregnancy or fibrocystic breast disease were excluded from the study. All subjects provided informed consent. Serum CA 15-3 levels were measured with a microparticle immunoassay. Differences were analysed by the Student’s t-test, analysis of variance, and the chi-squared test for statistical analysis. Two patients with steady-state SCD had experienced a prior cerebrovascular event. No difference was found among patients with or without painful crisis with respect to transfusion requirement, bone necrosis, hepatomegaly, seropositivity for hepatitis B or hepatitis C virus or the median value of alanine aminotransferase, bilirubin or creatinine peak levels (P > 0Æ05). All patients had been pretreated with an avarage of three substances (usually hydroxycarbamide, zinc and folic acid). There was no difference between the groups regarding the treatment with hydroxycarbamide (P > 0Æ05).
In both SCD groups and the metastatic breast cancer group, serum CA 15-3 concentrations were elevated when compared with controls (P < 0Æ001 for all). The subjects with metastatic breast cancer were found to have significantly higher serum CA 15-3 concentrations when compared with the patients with SCD (steady-state disease or painful crisis) (P < 0Æ001 for both). There was no difference between the serum CA 15-3 levels of the patients with SCD in steady state or painful crisis (P > 0Æ05) (Table I). Twenty-eight patients (82%) with steady-state SCD and 13 patients (76%) with SCD in painful crisis had CA 15-3 values >30 U/ml. This difference was not statistically significant (P > 0Æ05). When both groups of patients with SCD were compared with controls, the percentage of CA 15-3 levels ‡30 IU/l was significantly higher than patients with SCD (P < 0Æ001 for both). However, this percentage in the SCD groups was lower than patients with breast cancer (P < 0Æ001 for both) (Table I). Among the patients with SCD, serum CA 15-3 levels were not affected by the use of hydroxycarbamide. Possible explanations of our findings, based on the current literature are: (i) in the past two decades, it has been established that SCD is an inflammatory state with abnormal endothelial cell activation (Hebbel et al, 2004). Previous reports have shown that elevated CA 15-3 levels are associated with connective tissue diseases, such as systemic lupus erythematosus (SLE), and systemic sclerosis (Colomer et al, 1989; Valerio Marzano et al, 1998). One study showed that 7% of patients with SLE had serum CA 15-3 concentrations >40 U/ml (Colomer et al, 1989). Pathogenetic mechanisms that cause the elevation of CA 15-3 in patients with an inflammatory disease can also explain the elevation of CA 15-3 in both SCD groups in our study. (ii) In bone marrow differentiating cells, the increase in MUC1 expression
Table I. Cancer antigen 15-3 level by study group.
Parameters Median value, CA 15-3 level; IU/l (range) CA 15-3 ‡ 30 IU/l; no versus yes (%)
SCD (steady-statea) (n ¼ 34)
SCD (painful crisisb) (n ¼ 17)
Breast cancerc (n ¼ 15)
Controlsd (n ¼ 20)
53Æ39 ± 23Æ25 (7Æ3–104Æ8) 6/28 (82)
45Æ77 ± 21Æ19 (12Æ0–94Æ9) 4/13 (76)
268Æ15 ± 296 (39Æ2–1245) 0/15 (100)
13Æ81 ± 8Æ43 (9Æ4–33Æ7) 19/1 (5)
P-value 0Æ05a-b 0Æ05a,b
Superscript letters indicate the study groups on which statistical comparisons were carried out. SCD, sickle cell disease; CA, cancer antigen.
546
ª 2006 The Authors Journal Compilation ª 2006 Blackwell Publishing Ltd, British Journal of Haematology, 134, 544–549
Correspondence has been shown at the mRNA level (Rughetti et al, 2003). MUC1 was expressed strongly but transiently in early erythroid differentiation, however, it was absent in the circulating erythrocytes (Rughetti et al, 2003). Moreover, increased serum CA 15-3 has also been reported among patients with homozygous b-thalassaemia and sickle cellthalassaemia (Symeonidis et al, 2006). These observations may suggest that in SCD, the erythroid hyperplasia might express MUC1 glycoform similar to those expressed by carcinoma cells. Our data indicates that serum CA 15-3 levels may be elevated in individuals with SCD. The serum concentration of CA 15-3 should thus be considered when malignant disease is suspected. Can Boga1 Hakan Ozdogu1 Nurzen Sezgin2 Ebru Kizilkilic1 Zafer Koc3 Hakan Sakalli4 Defne Yalcintas1 Ilknur Kozanoglu1 1
Haematology Division, Department of Internal Medicine, 2Department
of Biochemistry, 3Department of Radiology, and 4Department of Medical
References Colomer, R., Ruibal, A., Genolla, J., Rubio, D., Del Campo, J.M., Bodi, R. & Salvador, L. (1989) Circulating CA 15-3 levels in the postsurgical follow-up of breast cancer patients and in non-malignant diseases. Breast Cancer Research and Treatment, 2, 123–133. Hayes, D.F., Zurawski, Jr, V.R. & Kufe, D.W. (1986) Comparison of circulating CA15-3 and carcinoembryonic antigen levels in patients with breast cancer. Journal of Clinical Oncology, 4, 1542–1550. Hebbel, R.P., Osarogiagbon, R. & Kaul, D. (2004) The endothelial biology of sickle cell disease: inflammation and a chronic vasculopathy. Microcirculation, 11, 129–151. Rughetti, A., Biffoni, M., Pierelli, L., Rahimi, H., Bonanno, G., Barachini, S., Pellicciotta, I., Napoletano, C., Pescarmona, E., Del Nero, A., Pignoloni, P., Frati, L. & Nuti, M. (2003) Regulated expression of MUC1 epithelial antigen in erythropoiesis. British Journal of Haematology, 120, 344–352. Symeonidis, A., Kouraklis-Symeonidis, A., Constandinidou, I., Solomou, E., Kougelou, S., Vassilakos, P. & Zoumbos, N. (2006) Increased CA-15.3 levels in the serum of patients with homozygous beta-thalassaemia and sickle cell/beta-thalassaemia. British Journal of Haematology, 133, 692–694. Valerio Marzano, A., Morabito, A., Berti, E. & Caputo, R. (1998) Elevated circulating CA 15.3 levels in a subset of systemic sclerosis with severe lung involvement. Archives of Dermatology, 134, 645.
Keywords: sickle cell disease, MUC1, cancer antigen 15-3, breast cancer.
Oncology, Faculty of Medicine, Baskent University, Ankara, Turkey. E-mail:
[email protected]
doi:10.1111/j.1365-2141.2006.06220.x
Chronic eosinophilic leukaemia with FIP1L1–PDGFRA fusion and T674I mutation that evolved from Langerhans cell histiocytosis with eosinophilia after chemotherapy
We have investigated the genetic abnormalities of a case of chronic eosinophilic leukaemia (CEL) with a unique course of disease progression. This 25-year-old man was initially diagnosed with Langerhans cell histyocytosis (LCH) due to multiple osteolytic lesions with infiltration of eosinophils and non-dysplastic eosinophilia in peripheral blood. Despite several courses of multidrug chemotherapy consisting of vinblastine, methotrexate, 6-mercaptopurine and cladribine, blastic eosinophils were found to have appeared in peripheral blood, bone marrow and pleural effusion 1 year later, and CEL was diagnosed. Administration of imatinib methylate caused transient remission of eosinophilia, but pleural effusion with eosinophilia recurred 2 months later.
Reverse transcription-polymerase chain reaction (RT-PCR) of bone marrow and pleural effusion cells at the CEL diagnosis successfully detected a FIP1L1–PDGFRA fusion gene, reported to be frequently found in hypereosinophilic syndrome (HES) (Cools et al, 2003a), and direct-sequencing analysis revealed that exon 10 of FIP1L1 and exon 12 of PDGFRA were fused, with the insertion of 30 bp of unknown origin (Fig 1A). Sequence analysis of exon 15 of PDGFRA using DNA prepared from the same samples showed germline sequence. However, pleural effusion cells at the time of relapse showed a T674I mutation within exon 15 of PDGFRA (Fig 1B). Next we tried to analyse FIP1L1–PDGFRA in the tissue of bone lesions with infiltration of eosinophils obtained prior
ª 2006 The Authors Journal Compilation ª 2006 Blackwell Publishing Ltd, British Journal of Haematology, 134, 544–549
547
